Convert HTTP_CLIENT_RECEIVE from UTF-16 to UTF-8

[dimath72]

Dear Friends ,

I have made a program that call an API and the answer is in UTF-16 and I cannot handle it. I have searched many hours on the forum but I did;t managed to fix it.

Can you propose something?

CALL METHOD P_HTTP_CLIENT->RECEIVE
    EXCEPTIONS
      HTTP_COMMUNICATION_FAILURE = 1
      HTTP_INVALID_STATE         = 2
      HTTP_PROCESSING_FAILED     = 3
      OTHERS                     = 4.

  P_P_CONTENT = P_HTTP_CLIENT->RESPONSE->GET_CDATA( ).

  CALL FUNCTION 'SCMS_STRING_TO_XSTRING'
    EXPORTING
      TEXT   = P_P_CONTENT
    IMPORTING
      BUFFER = GV_XML_STRING
    EXCEPTIONS
      FAILED = 1
      OTHERS = 2.


  CALL FUNCTION 'SMUM_XML_PARSE'
    EXPORTING
     XML_INPUT = GV_XML_STRING
   TABLES
     XML_TABLE = GWA_XML_DATA

Look the result that I take from server

And look the headers that I sent ....

Thanks a lot .....

[ThorstenHoefer]

Hi,

please check class cl_abap_conv_codepage, like

 cl_abap_conv_codepage=>create_in( CODEPAGE = 'UTF-16' )->convert( l_xstr ) 

[dimath72]

Can you please provide an example ?

[dimath72]

I am in version 740. I dont have it ...

[Ryan-Crosby]

Hi Dimitras,

Is the UTF-16 big endian or little endian?

Regards,

Ryan Crosby

[dimath72]

Actually, I don't know if it is Big or Little ....

[Ryan-Crosby]

Hi Dimitras,

You could try something like this for UTF-16LE (4102 for BE) - All parameters but one are mandatory so this is just a sample but you can tailor it however you want and see if it will work. Also, if this does not work there are some system -> external and external -> system format classes you can try: CL_ABAP_CONV_IN_CE, CL_ABAP_CONV_OUT_CE.

DATA: input TYPE xstring,
      output TYPE xstring,
      inused TYPE i,
      outused TYPE i,
      substed TYPE i,
      input_ends TYPE c,
      too_short TYPE c.
      

DATA(converter) = NEW cl_abap_conv_obj( incode = '4103' 
                                        outcode = '4110' ).
converter->convert( EXPORTING
                      inbuff             = input
                      outbufflg          = 10485760 "10MB buffer limit
                    IMPORTING
                      outbuff            = output
                      inused             = inused
                      outused            = outused
                      substed            = substed
                      input_ends_in_char = input_ends
                      outbuff_too_short  = too_short  ).

Regards,

Ryan Crosby

[dimath72]

Thanks a lot!

Correct answer.

You help me very much.

P.S)Thanks all for the help/

[Sandra_Rossi]

It's UTF-16LE (little endian) because of order <# (where # is in fact the null byte in your case). If order was #< that would be big endian.

Note that if you define an XML transformation to process the XML, you would use GET_DATA instead of GET_CDATA, and you wouldn't need to care about the encoding/codepage.